M R Underwood1, A G Barnett, M R Vickers. 1. MRC Epidemiology and Medical Care Unit, Wolfson Institute of Preventive Medicine, St. Bartholomew's, London. m.underwood@mds.qmw.ac.uk
Abstract
STUDY DESIGN: Postal questionnaire to individuals with back pain. OBJECTIVE: To assess the acceptability, validity, and reliability of two existing back pain outcome measures, the Roland-Morris Questionnaire and the Von Korff scales, modified to measure the preceding 4 weeks. SUMMARY OF BACKGROUND DATA: The ideal outcome measure for studies of low back pain and disability remains elusive. Most existing measures assess current pain and disability. Measuring these factors over a preceding 4-week period may be more appropriate. METHODS: Individuals with back pain identified in a community survey were asked to complete the modified questionnaires. Validity was assessed by comparison with the Medical Outcome Study Short Form 36 and two general comparator questions on self-reported pain and disability. Repeatability was assessed using retest questionnaires. RESULTS: Completed questionnaires were returned by 95 individuals with chronic back pain. The modified Roland-Morris Questionnaire and Von Korff pain and Von Korff disability scales were completed satisfactorily by 83 (87%), 89 (94%), and 87 (92%) participants, respectively. Mean scores of the modified measures changed significantly and in a predictable manner with increasing ratings of pain and disability. They also correlated with aspects of the Medical Outcome Study Short Form 36 questionnaire. Retest data suggest that these measures are repeatable. The modified Roland-Morris Questionnaire provided adequate analyzable data only if missing values were imputed, and it explained less of the variance in the comparator questions than the modified Von Korff scales. CONCLUSIONS: The modified Von Korff scales were completed easily and appear to be valid and repeatable in this format.
STUDY DESIGN: Postal questionnaire to individuals with back pain. OBJECTIVE: To assess the acceptability, validity, and reliability of two existing back pain outcome measures, the Roland-Morris Questionnaire and the Von Korff scales, modified to measure the preceding 4 weeks. SUMMARY OF BACKGROUND DATA: The ideal outcome measure for studies of low back pain and disability remains elusive. Most existing measures assess current pain and disability. Measuring these factors over a preceding 4-week period may be more appropriate. METHODS: Individuals with back pain identified in a community survey were asked to complete the modified questionnaires. Validity was assessed by comparison with the Medical Outcome Study Short Form 36 and two general comparator questions on self-reported pain and disability. Repeatability was assessed using retest questionnaires. RESULTS: Completed questionnaires were returned by 95 individuals with chronic back pain. The modified Roland-Morris Questionnaire and Von Korff pain and Von Korff disability scales were completed satisfactorily by 83 (87%), 89 (94%), and 87 (92%) participants, respectively. Mean scores of the modified measures changed significantly and in a predictable manner with increasing ratings of pain and disability. They also correlated with aspects of the Medical Outcome Study Short Form 36 questionnaire. Retest data suggest that these measures are repeatable. The modified Roland-Morris Questionnaire provided adequate analyzable data only if missing values were imputed, and it explained less of the variance in the comparator questions than the modified Von Korff scales. CONCLUSIONS: The modified Von Korff scales were completed easily and appear to be valid and repeatable in this format.